That doesn't explain western Europe, where for example France has gone from 1000 deaths per day to 10, with the same infection rate in the 2nd wave. If this was due to just testing, that would imply testing was off by 2 orders of magnitude, which would imply half of the population has already contracted the virus, which we know from antibody testing is not the case.
You see the same thing in New York, where fatalities per day has gone down from 1000 to 7, while cases per day is only down 90% from peak. Crowded cities it may be impossible to avoid exposure completely (which crazily may be a benefit!).
Rural US is where the fatality rate isn't going down much, which would make sense since there's higher chance of someone having had zero exposure to the virus and then getting a bunch of it at once by coming in contact with a sick person.
NOTE this is not based on any research other than some guesses as to how these numbers could line up. Do not use this as a source for anything!
> which would imply half of the population has already contracted the virus, which we know from antibody testing is not the case.
Antibodies disappear quickly so antibody studies aren't a perfect indicator of spread.
See for example:
> The UK study, meanwhile, suggested that antibodies could disappear completely within three months. The research tested more than 90 coronavirus patients for antibodies over 94 days. The results showed that three weeks after infection, just 60% of patients had developed a "potent" antibody response. In the weeks that followed, a mere 17% of those patients maintained that level of antibodies, The Guardian reported. In the other patients, antibody levels declined 23-fold or became undetectable within three months.
https://www.worldometers.info/coronavirus/country/france/
You see the same thing in New York, where fatalities per day has gone down from 1000 to 7, while cases per day is only down 90% from peak. Crowded cities it may be impossible to avoid exposure completely (which crazily may be a benefit!).
https://www.worldometers.info/coronavirus/usa/new-york/
Rural US is where the fatality rate isn't going down much, which would make sense since there's higher chance of someone having had zero exposure to the virus and then getting a bunch of it at once by coming in contact with a sick person.
NOTE this is not based on any research other than some guesses as to how these numbers could line up. Do not use this as a source for anything!